Quantifying genocide in Darfur
Implications and Limitations of a Recent Study
(“Death in Darfur,” published in Science magazine, September 15, 2006)
By Eric Reeves
September 16, 2006 — John Hagan of Northwestern University and Alberto Palloni of the University of Wisconsin (Madison) have, in the current issue of Science magazine, published a technically acute two-page Darfur mortality study:
“Death in Darfur”
John Hagan and Alberto Palloni
Science 15 September 2006 (313): pages 1578-1579
[DOI: 10.1126/science.1127397]
www.sciencemag.org
Hagan and Palloni conclude that, “the death toll in Darfur is conservatively estimated to be in the hundreds of thousands rather than tens of thousands of people.” This conclusion is important in light of continued inaccurate reporting on mortality in Darfur by a number of distinguished news organizations, including Reuters, which habitually refer to mortality in Darfur as numbering in “the tens of thousands.”
Hagan and Palloni offer a range of 170,000 to 255,000 deaths for 31 months of the Darfur conflict (which has now entered its 44th month), even as they acknowledge the possibility that “the death toll is much higher.”
But the severe limitations of the study are as significant as its highly circumscribed, if methodologically impeccable, achievements. Most notably, the study takes no account of two studies with critical implications for any meaningful assessment of global human mortality in Darfur:
[1] “Mortality Survey among Internally Displaced Persons and other affected populations in Greater Darfur, Sudan” (a 39-page UN World Health Organization-overseen study released in Khartoum, September 2005)
http://64.233.161.104/search?q=cache:gaLp-NnRyukJ:www.emro.who.int/sudan/pdf/CMS%2520Darfur%25202005%2520final%2520report_11%252010%252005.pdf+%22mortality+in+Darfur%22+khartoum+%22main+findings%22&hl=en&gl=us&ct=clnk&cd=7
(a report on the “Main Findings” was released by the World Health Organization in June 2005)
[2] “Documenting Atrocities in Darfur” (a highly ambitious and rigorous study organized by the Coalition for International Justice (CIJ), surveying Darfuri refugees along the Chad/Darfur border; September 2004, at http://www.state.gov/g/drl/rls/36028.htm)
The CIJ study is self-described as “based on semi-structured interviews with 1,136 randomly selected refugees at 19 locations in eastern Chad. [ ] The field data for the 1,136 interviews were compiled using a standardized data entry process that involved the collection and coding of detailed information from each refugee respondent’s set of answers. The researchers then used a statistical program to aggregate the data and analyze the results.”
The significance of the data from the Coalition for International Justice study was sufficiently compelling to have figured centrally in a previous morality by Professor Hagan, along with co-author Patricia Parker (University of Toronto), published in April 2005. The study concludes that approximately 400,000 people had died from disease, malnutrition, and violence (the website URL containing this study is no longer maintained; electronic text of the study has been retained by this writer and is available upon request).
Asked why the Coalition for International Justice (CIJ) data did not figure in the current study, Hagan is reported as saying: “the CIJ survey was a rich resource that simply did not fit the criteria for the current study” (“Darfur Death Toll Is Hundreds of Thousands Higher Than Reported, Study Says,” National Geographic News, September 14, 2006
http://news.nationalgeographic.com/news/2006/09/060914-darfur-deaths.html).
Whether the CIJ data “fit the criteria for the current study” or not, it is irresponsible of Hagan and Palloni not to acknowledge that this study exists and leads ineluctably to enormous totals for violent human destruction. Hagan, in his April 2005 study with Parker, conservatively estimated 143,000 deaths from violence alone on the basis of the CIJ data (and 254,000 deaths from “health causes” in displaced persons camps). This writer has argued that the CIJ data on violent mortality lead to a range of 220,000 to 270,000 violent deaths from February 2003 through April 2006 (see “QUANTIFYING GENOCIDE IN DARFUR: Current data for total mortality from violence, malnutrition, and disease,” April 28, 2006, at http://www.sudanreeves.org/Article102.html).
Just as troubling is the failure of Hagan and Palloni to account for the implications of the most recent UN World Health Organization (WHO) study (see above). Even the preliminary data released in June 2005 reveal that excess mortality from disease and malnutrition throughout greater Darfur likely exceeded 6,000 per month (see my “DARFUR MORTALITY UPDATE: Current data for total mortality from violence, malnutrition, and disease,” June 30, 2005, at
http://www.sudanreeves.org/Sections-index-req-viewarticle-artid-515-page-1.html). This WHO study reflected mortality during a period in which humanitarian access was relatively good, and significant improvements had been made in food delivery, primary medical care, water purification, and the provision of shelter; the excess Crude Mortality Rate (CMR) for Darfur (excess deaths per 10,000 of affected population per day) was approximately 0.6 at the time.
Since the period reflected in this WHO study (November 2004-May 2005), humanitarian conditions in the camps have deteriorated drastically, primarily because of the significant rise in insecurity that began in August/September 2005, and which has culminated in the current severe attenuation of humanitarian access and a collapse of operations in many areas. While no further mortality data will be forthcoming during this period of extreme and accelerating violence, there is strong reason to believe that the excess Crude Mortality Rate has increased very significantly throughout Darfur, as has the denominator for this rate (the total number of conflict-affected persons: 3.78 million for Darfur, according to the UN Office for the Coordination of Humanitarian Affairs [OCHA]; 350,000 in eastern Chad, using figures from OCHA, the UN High Commission for Refugees, and the UN World Food Program).
Because of their decision not to consider the implications of the June/September WHO study and data, Hagan and Palloni are obliged to offer a mortality range that represents only 31 of 43 months of conflict in Darfur. Twelve months—an entire year of deaths, from all causes—is completely unrepresented in this new study.
To be sure, the heterogeneity of data available from various mortality studies creates severe methodological difficulties. But excessive methodological fastidiousness will produce irresponsible results if there is not a full acknowledgement of the potential implications of statistically significant data, especially give the paucity of data currently available. Here Hagan and Palloni have failed, despite their professed goal of alerting the world to a severe under-reporting of human mortality in Darfur.
The authors are responsible for not doing enough to indicate the significance of their exclusion of Coalition for International Justice data, which Hagan acknowledges to be a “rich resource that simply did not fit the criteria for the current study.” More bewildering yet is their failure to acknowledge, even in a footnote, the existence of the 39-page World Health Organization-overseen study, “Mortality survey among Internally Displaced Persons and other affected populations in Greater Darfur, Sudan,” September 2005 (see above). Failure to take account of this important study, almost certainly the last to be undertaken in Darfur until conditions of much greater security prevail, leads to a further understating of excess mortality by many tens of thousands of human beings.
And again, as the authors clearly acknowledge, the mortality figures they offer are based on 31 months of the Darfur conflict, leaving a full year of mortality excluded from the 43 months of large-scale conflict to date.
Notable also is the absence of any data bearing on mortality in Chad, which has been continuous from the earliest phase of the Darfur conflict. To date, all mortality assessments have labored with the same handicap: there are simply no disaggregated data bearing on the significant mortality specific to Chad.
This writer believes that an assessment of all data available, from all sources, yields a figure of approximately 500,000 deaths from disease, malnutrition, and violence over the past 43 months of conflict in Darfur (see two-part mortality assessment, April 28 and May 13, 2006 at http://www.sudanreeves.org/Article102.html and http://www.sudanreeves.org/Article104.html).
This finding is finally not inconsistent with the concluding generalization offered by Hagan and Palloni (“we conclude that the death toll in Darfur is conservatively estimated to be in the hundreds of thousands rather than tens of thousands of people”). Indeed, Hagan is reported as declaring that,
“The death toll is likely much higher than [170,000 to 255,000], notes John Hagan, co-author of the report published in the September 15 Science. Their reported upper limit ‘likely increases to the 400,000 range if the further year of the conflict is estimated and if missing and presumed dead persons are included,’ he says.”
(“Darfur Dead Much Higher than Commonly Reported,” Scientific American, September 15, 2006,
http://www.sciam.com/article.cfm?chanID=sa003&articleID=00037EE7-D970-150A-997083414B7F0000)
But given the lack of even a gesture toward other important sources of mortality data, the Hagan/Palloni study has provided an easy occasion for the inevitable and evidently irresistible reductionism of North American and European news reporting, with correspondingly predictable headlines: Associated Press today reports on the study with the headline: “About 200,000 killed in Sudan’s Darfur region” (September 16, 2006).
It is a terribly grim irony that a study that is a self-described attempt to rectify news media misrepresentation of Darfur’s staggering morality total has in the end worked to solidify an untenably low estimate for this obscenely vast episode in ethnically-targeted human destruction.
NOTE 1:
In addition to the sources of systematic, statistically significant data, several reports on Darfur have offered a number of important generalized assessments, as well as suggestive anecdotal data. A good example of the latter appears in the January 2006 study from Physicians for Human Rights (which has undertaken a number of assessment missions to Darfur and eastern Chad). In “Darfur: Assault on Survival” (January 2006, http://www.phrusa.org/research/sudan/news_2006-01-11.html), PHR presents (on the basis of well-constructed interviews of carefully selected individuals in three representative locations) a shocking finding:
“Prior to the [military] attacks the 46 [Darfuri] men and women PHR interviewed had a total of 558 people in their households. Of these, 141 were ‘confirmed dead’—their deaths were witnessed or their bodies found—while 251 were ‘killed or missing’—meaning their whereabouts were unknown. The average household size [defined as ‘people who eat out of the same pot’] before the attacks was 12.1; after it was 6.7.”
This represents violent mortality of 45% for the family populations interviewed. We needn’t believe that the population sample in the report is statistically representative of Darfur as a whole to see that huge areas have suffered enormous violent human destruction. An epidemiological study of violence in West Darfur, published in the British medical journal The Lancet (October 1, 2004), found that 95% of those displaced in West Darfur had been violently displaced. Well over 2 million people have now been displaced, either internally within Darfur or into eastern Chad, suggesting that at least 2 million have been violently displaced. Even if the PHR figure of 45% mortality among the population of violently displaced persons overstates by 100% the violent mortality rate for Darfur as a whole, this would still imply statistically that approximately 450,000 human beings had died violent deaths.
Though this number is staggering, this is precisely the characterization made over two years ago by Asma Jahangir, then UN Special Rapporteur on extrajudicial, summary or arbitrary executions, who reported in late June 2004 that the “number of black Africans killed by Arab militias in the Darfur region of Sudan is ‘bound to be staggering’”:
“Ms. Jahangir said that during her visit, ‘nearly every third or fourth family’ she spoke to in the camps for internally displaced people (IDPs) within Darfur had lost a relative to the militias. ‘It’s very hard to say [accurately] how many people have been killed,’ she said, but interviews with IDPs indicated it would be ‘quite a large number. They are bound to be staggering.’” (UN News Centre, June 29, 2004)
NOTE 2:
There has been some highly informed reporting on the Hagan and Palloni study, by Scientific American and National Geographic News:
from Scientific American, September 15, 2006
SCIENCE NEWS
September 15, 2006
“Darfur Dead Much Higher than Commonly Reported”
http://www.sciam.com/article.cfm?chanID=sa003&articleID=00037EE7-D970-150A-997083414B7F0000
from National Geographic News, September 14, 2006
“Darfur Death Toll Is Hundreds of Thousands Higher Than Reported, Study Says”
http://news.nationalgeographic.com/news/2006/09/060914-darfur-deaths.html
* Eric Reeves is a professor at Smith College in Northampton, Massachusetts, and has published extensively on Sudan. He can be reached at [email protected]; website : www.sudanreeves.org